물류거점 데이터 예제 - 산점도(R)

앞서 만든 물류거점 데이터 예제 만들기 코드(https://ferti1969.tistory.com/140) 를 조금 더 진전시켜, 이번에는 각 관측치를 구분하는 이름이 붙은 산점도를 그리는 데까지 가 보기로 했다.

kind1 <- rep(1, 200)
base1 <- round(runif(200, min=1, max=10), digits=0)
emp1 <- round(runif(200, min=1, max=100), digits=0)
area1 <- round(runif(200, min=99, max=1500), digits=0)
volume1 <- round(runif(200, min=1, max=50000), digits=0)
arrival_car1 <- round(runif(200, min=1, max=500), digits=0)
shipment_car1 <- round(runif(200, min=1, max=500), digits=0)
data1 <- cbind(kind1, base1, emp1, area1, volume1, arrival_car1, shipment_car1)

kind2 <- rep(2, 75)
base2 <- round(runif(75, min=11, max=16), digits=0)
emp2 <- round(runif(75, min=50, max=500), digits=0)
area2 <- round(runif(75, min=99, max=3000), digits=0)
volume2 <- round(runif(75, min=100, max=75000), digits=0)
arrival_car2 <- round(runif(75, min=1, max=1000), digits=0)
shipment_car2 <- round(runif(75, min=1, max=1000), digits=0)
data2 <- cbind(kind2, base2, emp2, area2, volume2, arrival_car2, shipment_car2)

kind3 <- rep(3, 50)
base3 <- round(runif(50, min=17, max=20), digits=0)
emp3 <- round(runif(50, min=65, max=750), digits=0)
area3 <- round(runif(50, min=150, max=5000), digits=0)
volume3 <- round(runif(50, min=1000, max=100000), digits=0)
arrival_car3 <- round(runif(50, min=10, max=2500), digits=0)
shipment_car3 <- round(runif(50, min=10, max=2500), digits=0)
data3 <- cbind(kind3, base3, emp3, area3, volume3, arrival_car3, shipment_car3)

kind4 <- rep(4, 40)
base4 <- round(runif(40, min=21, max=23), digits=0)
emp4 <- round(runif(40, min=100, max=1000), digits=0)
area4 <- round(runif(40, min=500, max=10000), digits=0)
volume4 <- round(runif(40, min=5000, max=250000), digits=0)
arrival_car4 <- round(runif(40, min=50, max=3750), digits=0)
shipment_car4 <- round(runif(40, min=50, max=3750), digits=0)
data4 <- cbind(kind4, base4, emp4, area4, volume4, arrival_car4, shipment_car4)

kind5 <- rep(5, 35)
base5 <- round(runif(35, min=24, max=25), digits=0)
emp5 <- round(runif(35, min=150, max=5000), digits=0)
area5 <- round(runif(35, min=750, max=30000), digits=0)
volume5 <- round(runif(35, min=7500, max=500000), digits=0)
arrival_car5 <- round(runif(35, min=100, max=5000), digits=0)
shipment_car5 <- round(runif(35, min=100, max=5000), digits=0)
data5 <- cbind(kind5, base5, emp5, area5, volume5, arrival_car5, shipment_car5)

data <- rbind(data1, data2, data3, data4, data5)
colnames(data)

## [1] "kind1"         "base1"         "emp1"          "area1"        
## [5] "volume1"       "arrival_car1"  "shipment_car1"

colnames(data) <- c("kind", "base", "emp", "area", "volume", "arrival_car", "shipment_car")
colnames(data)

## [1] "kind"         "base"         "emp"          "area"         "volume"      
## [6] "arrival_car"  "shipment_car"

pairs(data[, 3:7])

matrix로 저장된 data를 data.frame으로 변경한 뒤, 각 관측치(=내 가상 데이터에서 각 행은 하나의 물류사업장임)를 구분하기 위한 no 변수를 만들자.

data <- as.data.frame(data)
data$no <- seq(1, 400)
head(data); tail(data)

##   kind base emp area volume arrival_car shipment_car no
## 1    1    9  95  319  41530         366          421  1
## 2    1    5  49  131  44620         414          375  2
## 3    1   10  22 1317  34445         481          348  3
## 4    1    3  49  492  33182         191          299  4
## 5    1    8   6  910   6967          98          431  5
## 6    1    4  68  521  10456         415           39  6

##     kind base  emp  area volume arrival_car shipment_car  no
## 395    5   25  500 18023 192750        1249         2048 395
## 396    5   25 3086 13007 182098         609         4760 396
## 397    5   25  890 12267 238236        1585         1329 397
## 398    5   24 1083 19216 320438         607          329 398
## 399    5   25 4097  6827 254434        1522         2523 399
## 400    5   25 2305 21118 288103        1277         1538 400

ggplot 패키지를 활용해 목표한 산점도를 그려 보자. 이 부분은 이 블로그(https://rfriend.tistory.com/72) 내용을 참고했다.

library("ggplot2")
ggplot(data = data, aes(x=emp, y=area)) +
  geom_point(shape = 19, size = 3, colour = "red") +
  ggtitle("Scatter Plot: emp vs. area with number") +
  geom_text(aes(label=no, vjust = -1, hjust = 0))